www.gradeup.

co

CONTROL SYSTEMS (FORMULA NOTES/SHORT NOTES)

Introduction

Open Loop Control Systems:

• In this system the output is not feedback for comparison with the input.

• Open loop system faithfulness depends upon the accuracy of input calibration.

C (s)
Transfer function, G ( s ) =
R (s)

When a designer designs, he simply design open loop system.

Closed Loop Control System:

It is also termed as feedback control system. Here the output has an effect on control action
through a feedback. Ex. Human being

Transfer Function:

C (s) G (s)
Transfer function = =
R (s) 1 + G (s) H (s)

Comparison of Open Loop and Closed Loop control systems:

Open Loop:

1. Accuracy of an open loop system is defined by the calibration of input.

2. Open loop system is simple to construct and cheap.

3. Open loop systems are generally stable.

2
www.gradeup.co

4. Operation of this system is affected due to presence of non-linearity in its elements.

Closed Loops:

1. As the error between the reference input and the output is continuously measured through
feedback. The closed system works more accurately.

2. Closed loop systems is complicated to construct and it is costly.

3. It becomes unstable under certain conditions.

4. In terms of performance the closed loop system adjusts to the effect of non-linearly present.

Transfer Function:

The transfer function of an LTI systems may be defined as the ratio of Laplace transform of output
of Laplace transform of input under the assumption.

Y (s)
G ( s) =
X (s)

• The transfer function is completely specified in terms of its poles and zeros and the gain factor.

• The TF function of a system depends on its elements, assuming initial conditions as zero and is
independent of the input function.

• To find a gain of system through transfer function put s = 0.

Example:

s+4 4
G ( s) = 2
Gain =
s + 6s + 9 9

If a step, ramp or parabolic response of TF is given, then we can find impulse response directly through
differentiation of that TF.

d
(Parabolic response) = Ramp response
dt

d
(Ramp response) = Step response
dt

d
(Step response) = Impulse response
dt

Block Diagram Reduction:

3
www.gradeup.co

4
www.gradeup.co

(GX1 ± X2)

Signal Flow Graphs:

• It is a graphical representation of control system.

• Signal flow graph of Block diagram.

Mason’s Gain Formula:

pk k
Transfer function =


pk → Path gain of kth forward path.

∆ = 1 – [Sum of all individual loops] + [Sum of gain products of two non-touching loops] – [Sum of
gain products of 3 non-touching loops] + …

 → Value of ∆ obtained by removing all the loops touching kth forward path as well as non-toughing
to each other.

Some Laplace and Z Transforms

5
www.gradeup.co

F ( s) f (nT ) F (z)
1 z
1 (nT )
s z −1
1 Tz
nT
s2 (z − 1)2
1 z
e−anT
s+a z − eaT
a
1−e − anT (
z 1 − eaT )
s ( s + a) (z − 1) ( z − e − aT
)
a (anT − 1 + e ) anT
 (  )
z z aT − 1 + e−aT + 1 − (1 + aT ) e−aT  

s 2
(s + a) a a ( z − 1)
2
(z − e )− aT

Laplace Transform:


F ( s ) = L f ( t ) = 0 f ( t ) e
−st
dt

Inverse Laplace Transform:

1 c +i st
f ( t ) = L−1 F ( s )  = e F ( s ) ds
2 c −i

Fourier Transform:


F ( j) = F f(t) =  f ( t ) e
− jt
dt

Inverse Fourier Transform:

1 
f ( t ) = F −1 F ( j)  = F ( j) e− jtd
2 

Star Transform:


F * ( s ) = L * f ( t ) =  f (iT ) e−siT
i =0

Z-Transform:


X ( z ) = Z x n  =  x n z−n
n =−

Inverse Z-Transform:

1
x n = Z−1 X ( z )  = X ( z ) zn−1dz
2j C

6
www.gradeup.co

Modified Z-Transform:


X ( z,m) = Z x n ,m =  x n + m − 1.z −n
n =−

Final Value Theorem:

x (  ) = lim sX (s) = lim (z − 1) X (z )

Initial Value Theorem:

x ( 0 ) = lim sX ( s )
s →

Euler’s Formula:

e j = cos ( ) + jsin ( )

Convolution:


( a * b)( t ) = − a ( ) b ( t − ) dT

Convolution Theorem:

L f ( t ) * g ( t )  = F ( s ) G ( s )
L f ( t ) g ( t )  = F ( s ) * G ( s )

Characteristic Equation:

A − I = 0
Av = v
wA = lw

Decibels:

dB = 20 log(C)

Unit Step Function:

0, t  0
u (t) = 
1, t  0

Unit Ramp Function: r(t) = t.u(t)

1 2
Unit Parabolic function: p (t) = t u (t )
2

Closed Loop Transfer Function:

7
www.gradeup.co

KGp ( s )
Hcl ( s ) =
1 + KGp ( s ) Gb ( s )

Open Loop Transfer Function:

Hol(s) = KGp(s)Gb(s)

Characteristics Equation:

F(s) = 1 + Hol

Time Response of 2nd order System:

Step input:

  1 − 2  
e−nt 
• C (t) = 1 − sin n 1 − 2 t  tan−1  
1 − 2     


  1 − 2 
e−nt 
• e (t) = sin nt  tan−1  
1 − 2  

 


  1 − 2 
e−nt  
• ess = lim sin nt  tan−1 
t →
1 − 2  

 


→  → Damping ratio; n → Damping factor

  1 (Under Damped):

  1 − 2 
e−nt
C (t) = 1 − sin nt  tan−1  
1 − 2    
  

 = 0 (Un damped):

C(t) = 1 – cosωnt

 = 1 (Critically damped) :

C ( t ) = 1 − e−nt (1 + nt )

  1 (Over damped) :

 
−  − 2 −1  nt
e  
C (t) = 1 −
(
2 2 − 1  −  2 − 1 )

8
www.gradeup.co

1
T=
( − )
2 − 1 n

Tundamped>Toverdamped>Tunderdamped > Tcritical damp

Time Domain Specifications:

−  1 − 2 
• Rise time tr =  = tan−1  
n 1 − 2   
 

n
• Peak time tp =
d

1−2
• Max over shoot % MP = e−n /  100

• Settling time, ts = 3T for 5% tolerance, ts = 4T for 2% tolerance

1 + 0.7
• Delay time, td =
n

(lnMp )
2
2
• Damping factor  2 =
2 + (lnMp )
2

2
• Time period of oscillations, T =
d

ts t  d
• No. of oscillations = = s
2 / d 2

• tr  1.5td tr = 2.2T

1
• Resonant peak, Mr = ;
2 1 − 2

n  r 
r = n 1 − 22    n  b
b  n  r

( )
1/2
• Bandwidth b = n 1 − 22 + 4 4 − 42 + 2

Static Error Coefficients:

• Step input:

9
www.gradeup.co

sR ( s )
ess = lim e ( t ) = lim E ( s ) = lim
t → s →0 s →0 1 + GH

1
ess = (Positional error)
1 + KP

KP = lim G ( s ) H ( s )
s →0

1
• Ramp input (t) = ess =
Kv

KP = lim S G ( s ) H ( s )
s →0

• Parabolic input (t2/2):

1
ess = , Ka = lim s2 G ( s ) H ( s )
Ka s →0

Type < input → ess = 

Type = input → ess finite

Type > input → ess = 0

A / A
• Sensitivity, S = sensitivity of A w.r.t. K.
K / K

• Sensitivity of overall T/F w.r.t. forward T/F G(s):

Open loop: S = 1

1
Closed loop: S =
1 + G (s) H (s)

• Minimum ‘S’ value preferable

G (s) H (s)
• Sensitivity of overall T/F w.r.t. feedback T/F H(s): S =
1 + G (s) H (s)

Stability RH Criterion:

• Take characteristic equation 1+G(s)H(s)=0

• All coefficients should have same sing.

• There should not be missing ‘s’ term. Term missed means presence of at least one +ve real part
root

• If characteristic equations contains either only odd/even terms indicates roots have no real part
and possess only image parts therefore sustained oscillations in response.

10
www.gradeup.co

• Row of all zeroes occur if

(a) Equation has at least one pair of real roots with equal image but opposite sign.

(b) has one or more pair of imaginary roots.

(c) has pair of complex conjugate roots forming symmetry about origin.

Position Error Constant:

Kp = lim G ( s )
s →0
Kp = lim G ( s )
z →1

Velocity Error Constant:

Kv = lim sG ( s ) = lim ( z − 1) G ( s )
s →0 z →1

Acceleration Error Constant:

2
Ka = lim s2G ( s ) = lim ( z − 1) G ( s )
s →0 z →1

General System Description:


y (t) = − g ( t,r ) x (r ) dr
Convolution Description:


y (t) = x (t )  h (t ) = − g ( ) h ( t − ) dt
Transfer Function Description:

Y (s) = H (s) X (s) Y (z ) = H (z ) X (z )

State Space Equations:

x ( t ) = Ax ( t ) + Bu ( t )

y ( t ) = Cx ( t ) + Du ( t )

Transfer Matrix:

−1
C sI − A B + D = H (s)

−1
C zI − A B + D = H (z)

Transfer Matrix Description:

11
www.gradeup.co

Y(s) = H(s)U(s)

Y(z) = H(z)U(z)

Mason’s Rule:

N
yout M 
M= =  k k
yin k =1 

State Space Methods:

1. General State Equation Solution:

t A ( t − )
x ( t ) = eAt − to x ( to ) + to e Bu (  ) d

n −1
x n = Anx 0 +  An−1−mBu n
m= 0

2. General Output Equation Solution:

t A ( t − )
y ( t ) = CeAt − to x ( t o ) + C  e Bu (  ) d + Du ( t )
to

n −1
y n = CAnx 0 +  CAn−1−mBu n + Du n
m= 0

3. Time Variant Equation Solution:

t
x ( t ) =  ( t, to ) x ( to ) + to  ( , to ) B (  ) u (  ) d x n =  n,no  x to  +   n,m + 1B mu m
m =no

4. Impulse Response Matrix:

C (  )  (r,  ) B (  ) if t  

G ( t,  ) = 

 if t  

CAk −1N if t  0
G n = 
 0 if k  0

Root Locus:

The magnitude Equation:

1 + KG ( s ) H ( s ) = 0

1 + KGH ( z ) = 0

The Angle Equation:

KG ( s ) H ( s ) = 180

12
www.gradeup.co

KGH ( z ) = 180

Number of Asymptotes:

Na = P – Z

Angle of Asymptotes:


k = (2k + 1)
P−Z

Breakaway Point Locations:

G (s) H (s) GH ( z )
= 0 or =0
ds dz

Controllers and Compensators:

PID:

Ki
D ( s ) = Kp + + Kds
s

T  z + 1  z − 1
D ( z ) = Kp + Ki  z − 1  + Ka  Tz 
2    

13